Narrow Results

In this paper, we investigate the problem of distributed optimization for second-order multi-agent systems with fixed-time flocking. The objective is to concurrently steer the agents towards a common velocity while optimizing the global objective function. To solve the problem, we first design a fixed-time estimator to predict the global gradient of the objective functions, then based on which, each agent is endowed with fixed-time tracking controller to track the optimal solution. Motivated by the fact that the relative velocity information is difficult to obtain accurately, the proposed algorithm is designed without using neighbors’ velocity. The upper bounds of the settling time are provided without relying on the initial states of the agents, only requiring the adjustment of parameters. The effectiveness of the proposed control protocol is also demonstrated through numerical simulations.

In this paper, we consider the output feedback event-triggered tracking control problem for a class of switched nonlinear multi-agent systems (MASs) with sensor faults and state constraints. For unknown nonlinear functions in the system, fuzzy logic systems are used to approximate them. To address sensor faults, sensor error compensation coefficients are designed to compensate for the errors caused by sensor faults. Meanwhile, state observer is designed to estimate the unmeasurable states in the system, providing necessary information for control design. Based on directed switching networks containing a directed spanning tree, an event-triggered output feedback controller is designed to enable the MAS to track the leader agent. By constructing a suitable barrier Lyapunov function, the stability of the system is analyzed, and it is proved that the consensus tracking can be achieved without violating the state constraints, the uniform boundedness of all the closed-loop system signals is ensured and the Zeno behavior can be eliminated. Finally, the effectiveness of the proposed algorithm is verified through a simulation example.

A cooperative team of agents may perform many tasks better than single agents. The question is how cooperation among self-interested agents should be achieved. It is important that, while we encourage cooperation among agents in a team, we maintain autonomy of individual agents as much as possible, so as to maintain flexibility and generality. This paper presents an approach based on bidding utilizing reinforcement values acquired through reinforcement learning. We tested and analyzed this approach and demonstrated that a team indeed performed better than the best single agent as well as the average of single agents.

The introduction of trust-based approaches in social scenarios modeled as multi-agent systems (MAS) has been recognized as a valid solution to improve the effectiveness of these communities. In fact, they make interactions taking place in social scenarios much fruitful as possible, limiting or even avoiding malicious or fraudulent behaviors, including collusion. This is also the case of multi-layered neural networks (NN), which can face limited, incomplete, misleading, controversial or noisy datasets, produced by untrustworthy agents. Many strategies to deal with malicious agents in social networks have been proposed in the literature. One of the most effective is represented by Eigentrust, often adopted as a benchmark. It can be seen as a variation of PageRank, an algorithm for determining result rankings used by search engines like Google. Moreover, Eigentrust can also be viewed as a linear neural network whose architecture is represented by the graph of Web pages. A major drawback of Eigentrust is that it uses some additional information about agents that can be a priori considered particularly trustworthy, rewarding them in terms of reputation, while the non pre-trusted agents are penalized. In this paper, we propose a different strategy to detect malicious agents which does not modify the real reputation values of the honest ones. We introduce a measure of effectiveness when computing reputation in presence of malicious agents. Moreover, we define a metric of error useful to quantitatively determine how much an algorithm for the identification of malicious agents modifies the reputation scores of the honest ones. We have performed an experimental campaign of mathematical simulations on a dynamic multi-agent environment. The obtained results show that our method is more effective than Eigentrust in determining reputation values, presenting an error which is about a thousand times lower than the error produced by Eigentrust on medium-sized social networks.

Despite a number of different numerical techniques existing for modeling the uptake of the nutrients, metabolism, maintenance, cell division and growth of bacteria population, none of them can be treated as a universally one. In this new model we have combined two techniques. The first one — agent-based — has been used for modeling the behavior of an individual bacterium. The agent objects (AOs) define generic features of the bacterium and the ways they interact with the environment and with the neighboring bacteria. The cellular automata are used for modeling the bacterial environment and represent communication layer for the agents, while a fixed two-dimensional grid defines the living space. The growth of the bacterial colony depends on the amount of free space in the closest neighborhood of individuals, which is required for reproduction, and on the availability of nutrients. We have matched the parameters of the model to demonstrate various growth structures developed by bacteria populations. We show that the patterns generated by the bacteria due to their collective behavior reflect the dynamical vitality of population and its fitness factor. We observe that the strongest populations self-organize in rod-like structures, which are reproduced in experimental microscopy images characteristic for bioflims and anthrax bacterial colonies.

The classic distributed computation is done by atoms, molecules or spins in vast numbers, each equipped with nothing more than the knowledge of their immediate neighborhood and the rules of statistical mechanics. These agents, 10²³ or more, are able to form liquids and solids from gases, realize extremely complex ordered states, such as liquid crystals, and even decode encrypted messages. We will describe a study done for a sensor-array "challenge problem" in which we have based our approach on old-fashioned simulated annealing to accomplish target acquisition and tracking under the rules of statistical mechanics. We believe the many additional constraints that occur in the real problem can be folded, step by step, into this stochastic approach. The results have applicability to other network management problems on scales where a distributed solution will be mandatory.

The statistical mechanics approach to wealth distribution is based on the conservative kinetic multi-agent model for money exchange, where the local interaction rule between the agents is analogous to the elastic particle scattering process. Here, we discuss the role of a class of conservative local operators, and we show that, depending on the values of their parameters, they can be used to generate all the relevant distributions. We also show numerically that in order to generate the power-law tail, an heterogeneous risk aversion model is required. By changing the parameters of these operators, one can also fine tune the resulting distributions in order to provide support for the emergence of a more egalitarian wealth distribution.

This paper investigates the leader-follower exponential consensus problem of a class of Lipschitz nonlinear multi-agent systems (MASs) with input saturation. Since each agent has nonlinear dynamics, the system is not asymptotically null controllable with bounded controls. Therefore, the widely-used low-gain feedback method for designing consensus protocols of MASs with input saturation can no longer work. Taking advantage of the stability theory of impulsive systems and features of the Laplacian matrix, and combining the properties of convex hull, a distributed impulsive consensus protocol is proposed. Still, the shape reference set is introduced to assess the attraction domain of leader–follower MASs. Finally, a numerical experiment validates the effectiveness of the proposed anti-saturation impulsive consensus algorithm.

In this paper, group consensus of second-order multi-agent systems with nonlinear dynamics is investigated. First, we design the distributed protocols for achieving group consensus, in which the strengths of the interactions among the agents are enhanced through tuning the coupling strengths. Further, taking the difference of the edges among agents into account, edge-based distributed protocols through tuning coupling weights of a fraction of edges are designed. Remarkably, only the edges of spanning tree in each group are pinned and the coupling strengths or weights of pinned edges are enhanced according to the updated laws. Both the types of distributed protocols are proved analytically and verified by numerical illustrations.

In this paper, we mainly investigate the character of consensus convergence speed of Multi-Agent Systems (MAS) with small-world communication network and the method of devising a speed-optimized small-world communication network in consensus problem based on genetic-algorithm (GA). It is found that, for a small-world communication network, the time to reach a consensus changes rapidly with the change of the number of long-range communication links and the agents which the long-range links connect. The convergence speed of consensus of MAS increases rapidly with the number of long-range links increasing. As we construct a small-world communication network for MAS with a smaller network size and fixed long-range links, we can optimize the long-range link configuration using GA methodology to obtain a small-world communication network with faster consensus speed for MAS.

This paper investigates group consensus for linear multi-agent systems with nonidentical dynamics. A novel adaptive event-triggered communication scheme is presented by using the stochastic sampling information, the event-triggered matrices and time-varying event-triggered parameters are introduced into event-triggered condition, where event-triggered parameters can be adjusted with the system dynamics evolving. The group consensus protocol is designed based on the neighboring agents information at event-triggered instants, then a new stochastic sampled-data dependent error model is constructed, some group consensus criteria in mean-square can be derived and the feedback matrices can also be obtained. Finally, two numerical examples are provided to illustrate the validity of the theoretical results.

This paper describes our approach to building a scalable, flexible agent-based architecture for imagery and geospatial processing. Central to this approach is the agent discovery and composition mechanism which scales to support networks with thousands of agents. The agent architecture implements over 100 imagery and geospatial processing agents based on the Java Advanced Imaging and OpenMap(TM) APIs. The agents are distributed over a Jini enabled network, and communicate with one another via JavaSpaces. We discuss our "atomic" approach in this paper: developing low-level processing agents that are used by application of specific agents. We discuss several concepts in this approach: agent lookup and discovery through traditional information retrieval techniques, the ability to rapidly prototype agents based on commercial software products, and a knowledge management approach that reuses prior processing approaches and results. We present results demonstrating the scalability of our agent discovery and composition mechanism to compare them with other traditional discovery mechanisms, and demonstrate how the discovery mechanism scales to support thousands of agents.

In a multi-agent system, an agent may utilize its idle time to assist other agents in the system. Intent recognition is proposed to accomplish this with minimal communication. An agent performing recognition observes the tasks other agents are performing and, unlike the much studied field of plan recognition, the overall intent of an agent is recognized instead of a specific plan. The observing agent may use capabilities that it has not observed. A conceptual framework is proposed for intent recognition systems. An implementation of the conceptual framework is tested and evaluated. We hypothesize that using intent recognition in a multi-agent system increases utility (where utility is domain specific) and decreases the amount of communication. We test our hypotheses using the domain of Cow Herding, where agents attempt to herd cow agents into team corrals. A set of metrics, including task time and number of communications, is used to compare the performance of plan recognition and intent recognition. In our results, we find that intent recognition agents communicate fewer times than plan recognition agents. In addition, unlike plan recognition, when agents use the novel approach of intent recognition, they select unobserved actions to perform. Intent recognition agents were also able to outperform plan recognition agents by consistently scoring more points in the Cow Herding domain. This research shows that under certain conditions, an intent recognition system is more efficient than a plan recognition system. The advantage of intent recognition over plan recognition becomes more apparent in complex domains.

Multi-agent systems can very naturally be viewed as computational organisations. For this reason, we believe organisational abstractions offer a promising set of metaphors and models that can be exploited in the analysis and design of such systems. To this end, the concept of role models is increasingly being used to specify and design multi-agent systems. However, this is not the full picture. In this paper we introduce three additional organisational concepts — organisational rules, organisational structures, and organisational patterns — and discuss why we believe they are necessary for the complete specification of computational organisations. In particular, we focus on the concept of organisational rules and introduce a formalism, based on temporal logic, to specify them. This formalism is then used to drive the definition of the organisational structure and the identification of the organisational patterns. Finally, the paper sketches some guidelines for a methodology for agent-oriented systems based on our expanded set of organisational abstractions.

With increasing requirements of distributed software systems, software agents are becoming a mainstream technology for software engineering and data management. Scalability and adaptability are two key challenges that must be addressed. In this work a new model is introduced for building large-scale distributed software systems with high dynamics, using a hierarchy of homogeneous agents that has the capability of service discovery. The performance of the agent system can be improved using different combinations of optimisation strategies. A modelling and simulation environment has been developed to aid the performance evaluation process. Two case studies are given and simulation results are included that show the impact of the agent mobility and the choice of performance optimisation strategies on the overall system performance.

Cooperative Information Systems (CIS) become relevant to integrate different kinds of systems so as to work collaboratively for a common goal. CIS are considered by nature as dynamic systems, and one of the most difficult problems is how to model and control multiple simultaneous interactions among agents in a friendly way. Consequently, expressiveness becomes a problem related to the representation so far, the similar systems cope neither with the problem of expressiveness nor with multiple interactions in a satisfactory way. It is proposed an integrated methodology based on Coloured Petri Nets (CPN) in order to model the interaction mechanism in a CIS and reduce the associated complexity in the representation of the dynamic of the system. The methodology integrates mainly: (a) the action basic loop in order to represent the system interactions and to model organization conversations, (b) the CPN in the interaction design and system simulation, (c) the communicative acts of FIPA (Foundation for Intelligent Physical Agents), included in the Agent Communication Language Specification.

How agents accomplish a goal task in a multi-agent system is usually specified by multi-agent plans built from basic actions (e.g. operators) of which the agents are capable. The plan specification provides the agents with a shared mental model for how they are supposed to collaborate with each other to achieve the common goal. Making sure that the plans are reliable and fit for the purpose for which they are designed is a critical problem with this approach. To address this problem, this paper presents a formal approach to modeling and analyzing multi-agent behaviors using Predicate/Transition (PrT) nets, a high- level formalism of Petri nets. We model a multi-agent problem by representing agent capabilities as transitions in PrT nets. To analyze a multi-agent PrT model, we adapt the planning graphs as a compact structure for reachability analysis, which is coherent to the concurrent semantics. We also demonstrate that one can analyze whether parallel actions specified in multi-agent plans can be executed in parallel and whether the plans can achieve the goal by analyzing the dependency relations among the transitions in the PrT model.

This article deals with the problem of dynamic role-playing in Multi-Agent organisations. The approach presented uses a formal specification notation and is based upon a formal framework which defines the concepts of role, interaction and organisation. Within this framework the problem of dynamic role-playing specification is related to the merging of specifications. The formal notation used composes Object-Z and Statecharts. The main features of this approach are: enough expressive power to represent Multi-Agents dynamic aspects, tools for specification analysis and mechanisms allowing the refinement of a high level specification into a low level specification which can be easily implemented. The last part of this paper presents an application with the specification of a reactive and cooperative MAS model named Satisfaction Altruism. An analysis of the specification validates the agents' behaviours.

In the past two decades, multi-agent systems have emerged as a new paradigm for conceptualizing large and complex distributed software systems. Even though there are many conceptual frameworks for using multi-agent systems, there is no well established and widely accepted method for the representation of multi-agent systems. We adapt a well-known formal model, predicate transition nets, to include the notions of dynamic structure, agent communication and coordination to address the representation problems. This paper presents a comprehensive methodology for modeling multi-agents based on the extensions. We demonstrate our modeling approach with an example. Several case studies on different application domains from our previous works are also discussed.

This paper presents a methodology for analyzing multi-agent systems modeled in nested predicate transition nets. The objective is to automate the model analysis for complex systems, and provide a foundation for tool development. We formally define the translation rules that translate the multi-agent model to an executable PROMELA model, and demonstrate the translation with an example.